Voucher-incentive contingency management (CM) interventions are efficacious in enhancing retention in treatment and reducing cocaine use. However, voucher procedures have not been implemented widely in community-based treatment programs. A lower-cost CM procedure, that provides opportunities to win prizes ranging in value from 1 dollar to 100 dollars, shows preliminary efficacy in applied treatment settings. Study I will compare directly the voucher and prize procedures. 120 individuals meeting DSM-IV criteria for cocain dependence will be randomly assigned to one of three conditions: (a) standard treatment plus CM in which subjects can earn up to 900 dollars in vouchers; (b) standard treatment plus CM with an expected probability of winning 300 dollars worth of prizes; or (c) a control condition-standard treatment without CM. In the two CM conditions, subjects will be reinforced for providing drug-free urine specimens and for complying with steps toward treatment goals. Drug use, severity of psychosocial problems, and HIV risk behaviors will be measured pre-treatment, during a 3-month treatment period, and throughout a 9-month follow-up period. Additionally, we will evaluate the cost-effectiveness of the interventions by assessing subjects' receipt of psychosocial and medical services, as well as criminal justice system involvement. Although the voucher system is expected to improve outcomes compared to standard treatment, it may not be significantly more efficacious, or cost-effective, than the prize system. Using the most efficacious and cost-effective treatment from Study I (e.g., vouchers or prizes), Study II will identify the active components of the CM intervention. 120 subjects will be randomly assigned to one of three conditions: (a) standard treatment plus CM, with vouchers or drawings contingent only upon submission of clean urine samples; (b) standard treatment plus CM, with vouchers or drawings contingent only upon completion of activities related to treatment goals; or (c) standard treatment without CM. Study III will assess the efficacy of CM when provided and managed entirely by clinical staff, thereby addressing transferability of the procedures. Following a staff training program, 240 subjects will be randomly assigned to one of two conditions: (a) standard treatment plus an efficacious CM intervention from Studies I and II or (b) standard treatment without CM. In Study III, all aspects of CM will be managed entirely by clinical staff in three sites. Together these studies will determine an efficacious and relatively simple CM treatment strategy, as well as test its effectiveness when delivered and managed by non-research staff.